start-vibing 3.0.1 → 3.0.2

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (2) hide show
  1. package/package.json +1 -1
  2. package/template/.claude/CLAUDE.md +64 -55
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "start-vibing",
3
- "version": "3.0.1",
3
+ "version": "3.0.2",
4
4
  "description": "Setup Claude Code agents, skills, and hooks in your project. Smart copy that preserves your custom domains and configurations.",
5
5
  "type": "module",
6
6
  "bin": {
package/template/.claude/CLAUDE.md CHANGED
@@ -49,35 +49,36 @@ If ONLY "Last Change" was modified but **significant** categorized source files
49
49
 
50
50
  ```
51
51
  .claude/
52
- ├── agents/ # 82 specialized agents in 14 categories
53
- ├── skills/ # 22 skill systems with cache
52
+ ├── agents/ # 6 active subagents (flat structure)
53
+ │ └── _archive/ # 82+ archived agents (unused, kept for reference)
54
+ ├── skills/ # 23 skill systems with cache
54
55
  ├── scripts/ # Validation scripts (validate-claude-md.ts)
55
56
  ├── config/ # Project-specific configuration
56
- ├── commands/ # Slash commands
57
+ ├── commands/ # Slash commands (feature, fix, research, validate)
57
58
  └── hooks/ # stop-validator.ts, user-prompt-submit.ts
58
59
  ```
59
60
 
60
61
  ---
61
62
 
62
- ## MCP Servers (MANDATORY FOR AGENTS)
63
+ ## MCP Servers
63
64
 
64
- All agents MUST use these MCP servers when applicable:
65
+ Available MCP servers and their usage:
65
66
 
66
- | Server | Purpose | Use In Agents |
67
- | --------------------- | ---------------------------------- | --------------------------------------- |
68
- | `context7` | Up-to-date library documentation | research, analyzer, tester |
69
- | `sequential-thinking` | Complex problem-solving | orchestrator, analyzer, final-validator |
70
- | `memory` | Persistent knowledge graph | domain-updater, commit-manager |
71
- | `playwright` | Browser automation (via MCP tools, persistent sessions via `--user-data-dir`) | ui-ux-reviewer |
72
- | `nextjs-devtools` | Next.js specific development tools | analyzer (Next.js projects) |
73
- | `mongodb` | MongoDB database operations | tester, security-auditor |
67
+ | Server | Purpose | Used By |
68
+ | --------------------- | ---------------------------------- | ---------------------------------------- |
69
+ | `context7` | Up-to-date library documentation | research-web agent, direct use |
70
+ | `sequential-thinking` | Complex problem-solving | Direct use for planning |
71
+ | `memory` | Persistent knowledge graph | domain-updater agent, direct use |
72
+ | `playwright` | Browser automation (`--user-data-dir` for persistent sessions) | Direct use for E2E testing |
73
+ | `nextjs-devtools` | Next.js specific development tools | Direct use in Next.js projects |
74
+ | `mongodb` | MongoDB database operations | Direct use for DB queries |
74
75
 
75
- ### Agent MCP Usage Rules
76
+ ### MCP Usage Rules
76
77
 
77
- - **research agent**: MUST use `context7` for library docs before recommending patterns
78
- - **analyzer agent**: SHOULD use `context7` to verify current API patterns
79
- - **domain-updater**: SHOULD use `memory` to persist patterns across sessions
80
- - **ui-ux-reviewer**: MUST use `playwright` MCP tools to verify UI implementations
78
+ - Use `context7` before implementing features with external libraries
79
+ - Use `memory` to persist patterns across sessions
80
+ - Use `playwright` for E2E testing (sessions persist via `--user-data-dir`)
81
+ - Plugins handle security (security-guidance hook) and types (typescript-lsp) automatically
81
82
 
82
83
  ---
83
84
 
@@ -99,18 +100,26 @@ Project-specific settings are in `.claude/config/`:
99
100
 
100
101
  ## Agent → Skill Mapping
101
102
 
102
- | Agent | Primary Skill | Secondary |
103
- | ---------------- | ------------------ | -------------------------------- |
104
- | analyzer | codebase-knowledge | - |
105
- | research | research-cache | codebase-knowledge |
106
- | documenter | docs-tracker | codebase-knowledge |
107
- | tester | test-coverage | - |
108
- | ui-ux-reviewer | ui-ux-audit | - |
109
- | security-auditor | security-scan | - |
110
- | quality-checker | quality-gate | - |
111
- | final-validator | final-check | ALL |
112
- | commit-manager | workflow-state | docs-tracker, codebase-knowledge |
113
- | domain-updater | codebase-knowledge | docs-tracker |
103
+ ### Active Subagents (6)
104
+
105
+ | Agent | Primary Skill | Secondary |
106
+ | ------------------ | ------------------ | ------------------ |
107
+ | research-web | research-cache | codebase-knowledge |
108
+ | documenter | docs-tracker | codebase-knowledge |
109
+ | domain-updater | codebase-knowledge | docs-tracker |
110
+ | commit-manager | workflow-state | docs-tracker |
111
+ | claude-md-compactor| - | - |
112
+ | tester-unit | test-coverage | - |
113
+
114
+ ### Plugins (replace archived agents)
115
+
116
+ | Plugin | Replaces Old Agent | Mechanism |
117
+ | ------------------ | ------------------- | ----------------- |
118
+ | security-guidance | security-auditor | PreToolUse hook |
119
+ | typescript-lsp | ts-strict-checker | LSP server |
120
+ | code-review | code-reviewer | /code-review cmd |
121
+ | frontend-design | ui-ux-reviewer | /frontend-design |
122
+ | commit-commands | commit-manager (PR) | /commit cmd |
114
123
 
115
124
  ---
116
125
 
@@ -121,16 +130,15 @@ Project-specific settings are in `.claude/config/`:
121
130
  1. **Read config** from `.claude/config/` for project specifics
122
131
  2. Read relevant skill SKILL.md file
123
132
  3. Check skill cache for existing data
124
- 4. Research if needed (web search)
133
+ 4. Research if needed (research-web agent or web search)
125
134
 
126
135
  ### After implementation:
127
136
 
128
137
  1. Update skill cache with changes
129
- 2. Run quality gates + record results
130
- 3. Security audit (if auth/data involved)
131
- 4. Final validation
132
- 5. **Update domains** via domain-updater agent (BEFORE commit, keeps git clean)
133
- 6. **Commit** via commit-manager agent (FINAL step)
138
+ 2. Run quality gates (bun run typecheck && lint && test)
139
+ 3. Security handled automatically by security-guidance plugin (PreToolUse hook)
140
+ 4. **Update domains** via domain-updater agent (BEFORE commit, keeps git clean)
141
+ 5. **Commit** via commit-manager agent (FINAL step)
134
142
 
135
143
  ---
136
144
 
@@ -208,14 +216,15 @@ export const usersApi = {
208
216
 
209
217
  ---
210
218
 
211
- ## VETO Power Agents
219
+ ## Enforcement Mechanisms
212
220
 
213
- These agents CAN and MUST stop the flow if rules are violated:
221
+ What blocks the flow if rules are violated:
214
222
 
215
- | Agent | Blocks When |
216
- | -------------------- | --------------------------------------------------- |
217
- | **security-auditor** | User ID from request, sensitive data, no validation |
218
- | **final-validator** | Any rule violated, tests failing, docs missing |
223
+ | Mechanism | Type | Blocks When |
224
+ | --------------------------- | -------------- | --------------------------------------------------- |
225
+ | **security-guidance plugin** | PreToolUse hook | Sensitive data exposure, missing validation, OWASP |
226
+ | **stop-validator hook** | Stop hook | Branch not main, uncommitted changes, docs missing |
227
+ | **quality gates** | Manual / CI | Typecheck, lint, or test failures |
219
228
 
220
229
  ---
221
230
 
@@ -223,15 +232,15 @@ These agents CAN and MUST stop the flow if rules are violated:
223
232
 
224
233
  All implementations MUST:
225
234
 
226
- - [ ] Pass typecheck (command from config)
227
- - [ ] Pass lint (command from config)
228
- - [ ] Pass unit tests (command from config)
229
- - [ ] Pass build (command from config)
235
+ - [ ] Pass typecheck (`bun run typecheck`)
236
+ - [ ] Pass lint (`bun run lint`)
237
+ - [ ] Pass unit tests (`bun run test`)
238
+ - [ ] Pass build (`bun run build`)
230
239
  - [ ] Use Playwright MCP for browser testing when needed
231
- - [ ] Have documentation in `docs/`
232
- - [ ] Be security audited
240
+ - [ ] Have documentation updated (documenter agent)
241
+ - [ ] Security validated by security-guidance plugin (automatic via hook)
233
242
  - [ ] Be committed with conventional commits
234
- - [ ] Have domains updated with session learnings
243
+ - [ ] Have domains updated with session learnings (domain-updater agent)
235
244
 
236
245
  ---
237
246
 
@@ -385,18 +394,18 @@ The **domain-updater** runs BEFORE commit-manager to ensure git stays clean.
385
394
 
386
395
  ### Trigger
387
396
 
388
- Runs automatically:
397
+ Runs in the workflow:
389
398
 
390
- 1. After final-validator approves in any workflow
399
+ 1. AFTER implementation and quality gates pass
391
400
  2. BEFORE commit-manager (so changes are included in commit)
392
- 3. Stop hook blocks session end until domain-updater executes (if files were modified)
401
+ 3. Stop hook blocks session end until domains are updated (if source files were modified)
393
402
 
394
403
  ### Workflow Order
395
404
 
396
405
  ```
397
- final-validator → domain-updater → commit-manager → complete-task
398
- ↑ ↑
399
- (updates domains) (commits all + archives)
406
+ implement → quality gates → domain-updater → commit-manager → complete-task
407
+ ↑ ↑
408
+ (updates domains) (commits all changes)
400
409
  ```
401
410
 
402
411
  ### Configuration